JP5714076B2 - Auditory saliency evaluation apparatus, auditory saliency evaluation method, and program - Google Patents

Description

  The present invention relates to an auditory saliency evaluation apparatus, an auditory saliency evaluation method, and a program that can be used for evaluation of auditory saliency that represents the degree of attention to sound.

  Auditory saliency is an important characteristic that expresses the degree of attention of sound. A paired comparison method has been used as a method for evaluating the auditory saliency of an arbitrary sound segment (Non-Patent Document 1, particularly FIG. 2). Further, Non-Patent Document 2 is known as a method for evaluating the saliency of a target sound by using the characteristics of ease of beats by the conspicuousness of sound.

Kayser et al., “Mechanisms for Allocating Auditory Attention: An Auditory Saliency Map,” Current Biology, Vol. 15, pp.1945-1947 (2005) Chon & McAdams, “Investigation of timbre saliency, the attention-capturing quality of timbre,” J. Acoust. Soc. Am., Vol.131 (4), pp.3433, (2012)

  In the rating method, the listener assigns a numerical value to the sound to be evaluated (target sound) according to the strength of the subjective saliency, and the saliency is evaluated by the magnitude of the numerical value. However, the evaluation value obtained by this method depends on the configuration of the target sound set used for the measurement. For this reason, it is difficult to compare numerical values obtained with different sound sets.

  In the pair comparison method, a plurality of target sound pairs are prepared for a target target sound set. For each target sound pair, the listener is made to judge which sound is more prominent. By statistically processing the result, the saliency of the target sound is ranked and digitized. Since this method requires evaluation for each sound pair, if the number of target sound targets increases, the number of target sound pairs required increases explosively, and within the range of realistic time and effort. This makes it difficult to conduct experiments. Moreover, since the numerical value is based on relative evaluation in a given set of target sounds, it is difficult to compare numerical values obtained in different sets as in the rating method. Non-Patent Document 2 uses the repetition of two types of target sounds as a sound string and evaluates the relative saliency of the target sound pairs, thus solving the problem seen in the above-described pair comparison method. I can't do it.

  As described above, the conventional method has a problem that the evaluation value of the saliency depends on the sound set used for the evaluation. Therefore, an object of the present invention is to provide an auditory saliency evaluation apparatus capable of quantitatively evaluating the auditory saliency of an arbitrary target sound without depending on the sound set used for evaluation.

  The auditory saliency evaluation apparatus of the present invention includes a sound presentation unit, an input information acquisition unit, and an evaluation value calculation unit.

  The sound presenting unit presents a sound string composed of a plurality of first sounds and a plurality of second sounds spaced apart from each other by a time interval. The input information acquisition unit acquires a series of beat information of a person who beats while listening to the presented sound string. The evaluation value calculation unit calculates the auditory saliency level based on the ratio of beat information corresponding to one of the two sounds in the series of beat information within a predetermined time interval.

  According to the auditory saliency evaluation apparatus of the present invention, the auditory saliency of an arbitrary target sound can be quantitatively evaluated without depending on the sound set used for the evaluation.

The figure explaining the structural example of the sound string shown to a listener by evaluation experiment of auditory saliency. The figure which shows the auditory saliency level for every kind of target sound which is an evaluation experiment result. The figure which compares and shows the experimental result of the conventional method (pair comparison method) and an invention method. The figure which shows the correlation of the experimental result of a conventional method (pair comparison method) and an invention method. 1 is a block diagram illustrating a configuration of an auditory saliency evaluation apparatus according to Embodiment 1. FIG. 5 is a flowchart showing the operation of the auditory saliency evaluation apparatus according to the first embodiment. The block diagram which shows the structure of the auditory saliency evaluation apparatus of Example 2. FIG. 10 is a flowchart showing an operation of the auditory saliency evaluation apparatus according to the second embodiment.

  Hereinafter, embodiments of the present invention will be described in detail. In addition, the same number is attached | subjected to the structure part which has the same function, and duplication description is abbreviate | omitted.

<Principle>
Non-patent document 2 also uses the characteristics of the ease of beats by the conspicuousness of the sound, where two different sounds are presented alternately and the listener beats in accordance with either sound. In general, it is a characteristic that a listener can easily beat a beat in accordance with a more conspicuous (highly noticeable) sound.

  On the other hand, the ease of engraving beats also depends on the volume (sound pressure level). If the same sound is presented alternately at different sound pressure levels, the listener is generally more likely to beat the loud sound. In the present invention, the saliency of the target sound is quantified by combining these two characteristics.

<Experimental conditions>
A sound string used in the evaluation experiment will be described with reference to FIG. FIG. 1 is a diagram for explaining a configuration example of a sound string presented to a listener in an auditory saliency evaluation experiment. First, a target sound (Target) 5 and a reference sound (Reference) 6 are prepared. A plurality of target sounds 5 shown in FIG. 1 are distinguished from each other by adding branch numbers -1, 2, 3,. Similarly, a plurality of reference sounds 6 shown in FIG. 1 are distinguished by adding branch numbers -1, 2, 3,. In the example of FIG. 1, the target sound 5 and the reference sound 6 are alternately and repeatedly arranged. An arbitrary time interval 7 is provided between the sounds. A plurality of time intervals 7 shown in FIG. 1 are distinguished from each other by adding a branch numeral -1, 2, 3,. In the example of FIG. 1, a time interval 7-1 is inserted after the target sound 5-1, a reference sound 6-1 is arranged after the time interval 7-1, and a time interval 7-2 after the reference sound 6-1. Is inserted after the time interval 7-2, and the time interval 7 is always inserted between the target sound 5 and the reference sound 6 that are alternately arranged, It is composed. In the example of FIG. 1, the reference sound 6 is presented after the target sound 5, but the order may be reversed. Further, it is not always necessary to present the target sound 5 and the reference sound 6 alternately such as the reference sound 6-1, the target sound 5-2, the reference sound 6-2, the target sound 5-3,. As long as the person can grasp the flow of the sound and cut the beat, for example, the reference sound 6-1, the reference sound 6-2, the target sound 5-1, the reference sound 6-3, the reference sound 6-4, and the target sound 5- One of the sounds may be presented continuously, such as 2,. In short, the target sound 5 and the reference sound 6 may not be presented randomly, but may be presented repeatedly with a time interval 7. The target sound 5 and the reference sound 6 may have more general names. For example, the target sound 5 and the reference sound 6 may be referred to as the first sound and the second sound. The target sound 5 may be assigned to the first sound, the reference sound 6 may be assigned to the second sound, the target sound 5 may be assigned to the second sound, and the reference sound 6 may be assigned to the first sound. The sound string is not limited to the example of FIG. 1, and may be composed of a plurality of first sounds and a plurality of second sounds that are spaced apart from each other.

  A plurality of types of target sounds 5 are prepared. The reference sound 6 is an arbitrary sound common to a plurality of types of target sounds 5. Hereinafter, experimental results using pink noise as the reference sound 6 will be described. The reference sound 6 is not limited to pink noise, and may be any sound as long as it is common among the target sounds 5. Note that the duration (X [ms] in FIG. 1) and the presentation interval (Y, Z [ms] in FIG. 1) of the target sound 5 and the reference sound 6 used in the evaluation experiment are 300 [ms] and 100 [ However, the present invention is not limited to this, and the duration and the presentation interval of the target sound and the reference sound have a degree of freedom and can be arbitrarily set within a predetermined range. Note that the time intervals Y [ms] and Z [ms] in FIG. That is, Y and Z are of such a length that the reference sound and the target sound can be heard as separate sounds, and the input information input from the listener by the input information acquisition unit 14 to be described later is either sound. It is sufficient that the time interval is such that it can be identified whether or not it is tapped.

  Further, the lengths of the target sound 5 and the reference sound 6 (X [ms] in FIG. 1) do not need to be completely the same, but are desirably long enough to be heard as substantially the same length. This is because if there is a difference in the length of the sound, there is a possibility that tapping will be performed due to the difference in length, and there is a possibility that an index that accurately reflects the saliency of the sound itself may not be obtained. Because there is.

  The listener listens to the sound string 8 presented repeatedly, presses the button in accordance with the sound that is easier to cut, and beats (tap) the beat. At this time, it can be said that the tapped sound is noticeable (highly noticeable). On the other hand, as described above, since it is easy to beat the sound with a higher sound pressure level, the sound to be tapped is switched by increasing or decreasing the sound pressure level of the reference sound 6. Can be considered. Therefore, when the target sound 5 is tapped, the sound pressure level of the reference sound 6 is increased, and when the target sound 6 is tapped, the sound pressure level of the reference sound 6 is decreased. By repeating this operation, it is possible to finally find the sound pressure level of the reference sound 6 such that the target sound 5 and the reference sound 6 are tapped at an equal ratio (a ratio of 50% each). Note that it is not essential to use the case where two sounds have the same probability of being tapped (50%), and the probability that one of the sounds is tapped (for example, the probability that the target sound is tapped is 60%). , 70%, etc.) may be set as a reference.

  For the target sound 5 having high saliency, it is considered that the sound pressure level of the reference sound 6 required to reach the above reference is relatively high. Similarly, for the target sound 5 with low saliency, it is considered that the sound pressure level of the reference sound 6 required to reach the reference is relatively low. For this reason, the sound pressure level of the reference sound when the target sound 5 and the reference sound 6 are tapped at a predetermined ratio (for example, 50% each) can be used as an index indicating saliency. Therefore, this sound pressure level is called an auditory saliency level. This auditory saliency level is determined independently for each of the plurality of types of target sounds 5. That is, by using the sound pressure level as the auditory saliency level, saliency can be expressed as an index independent of the set of the target sound 5. In addition to using the sound pressure level as an index, this experiment can be performed with the sound pressure level of the reference sound 6 fixed, and the probability of the target sound 5 being selected can be used as the saliency index.

<Experimental result>
Hereinafter, the results of the evaluation experiment will be described with reference to FIG. FIG. 2 is a diagram showing an auditory saliency level for each type of target sound 5 as an evaluation experiment result. The vertical axis of the graph in FIG. 2 represents the sound pressure level (auditory saliency level) of the reference sound 6, and the horizontal axis represents the number of the target sound 5 used in the experiment. FIG. 2 shows 1st: BEEP, 2nd: BIRD, 3rd: CHIRP, 4th: CRYING, 5th: DOG, 6th: LAUGHTER, 7th: PHONE, 8th: SCRATCH, 9th: TONE, 10 No .: Shows experimental results of 10 kinds of target sounds 5 of WHITE NOISE. The 11th PINK NOISE (pink noise) is the reference sound 6.

<Comparison between conventional method and invention method>
Hereinafter, the conventional method and the inventive method will be compared with reference to FIGS. FIG. 3 is a diagram comparing experimental results of the conventional method (paired comparison method) and the inventive method. In the conventional pair comparison method, 10 types of sound pairs used as the target sound 5 this time are prepared, and which sound is significant for each target sound pair is determined, and the obtained data is On the other hand, the scale of saliency was obtained by Thurston's paired comparison method. FIG. 4 shows the correlation between the experimental results of the conventional method (paired comparison method) and the inventive method. As shown in FIG. 4, the correlation coefficient between the inventive method and the conventional method is 0.79, and it can be seen that there is a correlation between the two. That is, when the same sound set is used, it can be seen that the auditory saliency level obtained by the inventive method has an evaluation value highly correlated with the conventional method.

  Hereinafter, the auditory saliency evaluation apparatus according to the first embodiment will be described with reference to FIGS. 5 and 6. FIG. 5 is a block diagram showing the configuration of the auditory saliency evaluation apparatus 1 of this embodiment. FIG. 6 is a flowchart showing the operation of the auditory saliency evaluation apparatus 1 of the present embodiment. As shown in FIG. 5, the auditory saliency evaluation apparatus 1 according to the present embodiment includes a storage unit 11, a control unit 12, a sound presentation unit 13, an input information acquisition unit 14, and an evaluation value calculation unit 15. . Hereinafter, the operation of each component will be described.

<Storage unit 11>
The storage unit 11 stores a target sound 5 and a reference sound 6. The reference sound 6 is a sound to be compared, and the target sound 5 is a sound to be evaluated. For example, the reference sound 6 is pink noise or the like, and the same sound prepared in advance is used regardless of the type of the target sound 5. The target sound 5 and the reference sound 6 may not be stored in the storage unit 11 in advance, and may be input from the outside, for example. In this case, the storage unit 11 can be omitted.

<Control unit 12>
The control unit 12 generates a sound string composed of a plurality of first sounds and a plurality of second sounds that are spaced over a predetermined time. At this time, the control unit 12 controls the number of repetitions of the first sound and the second sound (S12). More specifically, the control unit 12 includes a sound string 8 (a plurality of target sounds 5 having a predetermined time length and a plurality of reference sounds having a predetermined time length) presented in the sound presenting unit 13 during a predetermined time interval. 6 and a regular string of time intervals 7 inserted between them is generated. At this time, the control unit 12 controls the number of repetitions for generating the target sound 5 and the reference sound 6 (S12). In this example, the target sound 5 and the reference sound 6 are stored in the storage unit 11 in advance. The control unit 12 reads the target sound 5 and the reference sound 6 from the storage unit 11, and generates a sound string 8 by combining these sounds while inserting the time interval 7.

<Sound presentation unit 13>
The sound presentation unit 13 presents the sound string 8 generated by the control unit 12 (S13). In other words, the sound presenting unit 13 presents a sound string composed of a plurality of first sounds and a plurality of second sounds spaced apart from each other (S13). The sound presentation unit 13 includes, for example, an amplifier and a speaker or an amplifier and an earphone. Thus, the sound string 8 including the target sound 5 and the reference sound 6 generated by the control unit 12 is presented to a person who is a listener.

<Input information acquisition unit 14>
The input information acquisition unit 14 acquires a series of beat information of a person (listener) who beats while listening to the presented sound string (S14). More specifically, the input information acquisition unit 14 acquires, as a series of beat information, information on the time at which the listener beats (tapped) while listening to the sound string 8 repeatedly presented from the sound presenting unit 13. (S14). For example, the listener performs an operation of pressing (tapping) a button in accordance with a sound that easily beats. In this case, the input information acquisition unit 14 acquires time-series information at the time when the button is pressed (tapped) as a series of beat information (also referred to as input information).

<Evaluation Value Calculation Unit 15>
The evaluation value calculation unit 15 is auditory based on the ratio of the beat information corresponding to one of the two sounds (first and second sounds) in the series of beat information within the predetermined time interval. A saliency level is calculated (S15). Specifically, the evaluation value calculation unit 15 detects whether each of the series of beat information acquired by the input information acquisition unit 14 corresponds to the target sound 5 or the reference sound 6. Then, the ratio of the beat information corresponding to the target sound 5 in the entire series of beat information within the predetermined time interval is obtained, and this is output as an auditory saliency level (audience saliency evaluation value). The higher the auditory saliency level, the higher the saliency of the target sound 5, and the lower the auditory saliency level, the lower the saliency of the target sound 5. In other words, it indicates that the saliency of the reference sound 6 decreases as the auditory saliency level increases, and the saliency of the reference sound 6 increases as the auditory saliency level decreases.

  A series of input information (beat information) in a predetermined time interval instead of a ratio of input information (beat information) corresponding to the target sound 5 in a series of input information (beat information) in a predetermined time interval The ratio of input information (beat information) corresponding to the reference sound 6 may be used as the auditory saliency level. However, in this case, the higher the auditory saliency level is, the lower the saliency of the target sound 5 is, and the lower the auditory saliency level is, the higher the saliency of the target sound 5 is. In other words, the higher the auditory saliency level, the higher the saliency of the reference sound 6, and the lower the auditory saliency level, the lower the saliency of the reference sound 6. In any case, when the auditory saliency level is medium (around 50%), it means that the saliency of the reference sound and the target sound is the same (that is, there is no difference in saliency).

  Note that any sound can be used as the target sound 5 and the reference sound 6 in the present invention. For example, not only two sounds having different frequencies, periods, or sound pressures but also sounds having exactly the same (same frequency and same sound pressure) may be used. When the target sound 5 and the reference sound 6 are the same sound, the listener cannot distinguish between the target sound 5 and the reference sound 6, and therefore the target sound 5 and the reference sound 6 are assumed to have the same ratio of tapping. Is done. In this case, the level of auditory saliency obtained is moderate (around 50%). Therefore, an auditory saliency level of around 50 means that there is almost no difference in saliency between the target sound and the reference sound.

  In addition, the sound pressures of the target sound 5 and the reference sound 6 are not basically limited, but avoid a sound pressure that is so small that it cannot be heard or a sound pressure that is so high that it causes hearing loss, and can be heard without difficulty. It is desirable to keep it on.

  Hereinafter, with reference to FIG. 7 and FIG. 8, the auditory saliency evaluation apparatus according to the second embodiment in which a part of the first embodiment is changed will be described. FIG. 7 is a block diagram showing the configuration of the auditory saliency evaluation apparatus 2 of the present embodiment. FIG. 8 is a flowchart showing the operation of the auditory saliency evaluation apparatus 2 of the present embodiment. As shown in FIG. 7, the auditory saliency evaluation apparatus 2 of the present embodiment includes a storage unit 11, a control unit 22, a sound presentation unit 13, an input information acquisition unit 14, and an evaluation value calculation unit 25. Consists of including. The difference between the first embodiment and the present embodiment is that the control unit 12 in the first embodiment is changed to the control unit 22 in the present embodiment, and the evaluation value calculation unit 15 in the first embodiment calculates the evaluation value in the present embodiment. It is only the point which is changed to the part 25. Only the configuration different from the first embodiment will be described below.

<Control unit 22>
The control unit 22 of the present embodiment controls the generation of the sound string 8 and the number of repetitions as well as the control unit 12 of the first embodiment, and at the same time executes the sound pressure level control. Specifically, the control unit 22 has a predetermined ratio of beat information corresponding to one of the two sounds (first and second sounds) in a series of beat information within a predetermined time interval. If the ratio is larger than the ratio, the sound pressure level of the one sound is decreased (or the sound pressure level of the other sound is increased so that the sound pressure level of one sound is relatively decreased), and within a predetermined time interval When the ratio of beat information corresponding to one of the two sounds (first and second sounds) in the series of beat information is smaller than a predetermined ratio, the sound pressure level of the one sound Is increased (or the sound pressure level of the other sound is decreased so that the sound pressure level of one sound is relatively increased) (S22). For example, when “information indicating that many target sounds 5 have been tapped”, which will be described later, is input from the evaluation value calculation unit 14, the control unit 22 sets the sound pressure level of the reference sound 6 to be higher than the current sound pressure level. Control is made high (large), and when “information indicating that many reference sounds 6 have been tapped” is input, the sound pressure level of the reference sound 6 is controlled to be lower (smaller) than the current sound pressure level. . The sound presenting unit 13 presents the sound string 8 generated by the control unit 22 controlling the sound pressure level during a predetermined time interval (S13).

<Evaluation Value Calculation Unit 25>
The evaluation value calculation unit 25 calculates a value corresponding to the sound pressure level of one sound when the proportion of the beat information corresponding to one sound in the series of beat information within the predetermined time interval is equal to the predetermined ratio. An audio saliency level is output (S25).

  More specifically, first, the evaluation value calculation unit 25 detects whether the input information input from the input information acquisition unit 14 corresponds to the target sound 5 or the reference sound 6. Then, the ratio of the input information corresponding to the target sound 5 in the input information series within the predetermined time interval is obtained. Up to this point, the process is the same as in the first embodiment. Next, the evaluation value calculation unit 25 outputs “information indicating that many reference sounds 6 have been tapped” to the control unit 22 when the obtained ratio is smaller than the predetermined ratio. On the other hand, the evaluation value calculation unit 25 outputs “information indicating that many target sounds 5 have been tapped” to the control unit 22 when the obtained ratio is greater than the predetermined ratio. Further, when the calculated ratio is the same as the predetermined ratio, the evaluation value calculation unit 25 acquires the sound pressure level of the reference sound 6 at that time from the control unit 22 and determines the sound pressure level as auditory saliency. It outputs as a level (evaluation value of auditory saliency) (S25).

  In addition, instead of the ratio of the input information corresponding to the target sound 5 in the input information sequence in the predetermined time interval, the input information corresponding to the reference sound 6 in the input information sequence in the predetermined time interval May be used. In this case, the evaluation value calculation unit 25 outputs “information indicating that many reference sounds 6 have been tapped” to the control unit 22 when the obtained ratio is greater than the predetermined ratio. On the other hand, when the calculated ratio is smaller than the predetermined ratio, the evaluation value calculation unit 25 outputs “information indicating that many target sounds 5 have been tapped” to the control unit 22. Further, when the calculated ratio is the same as the predetermined ratio, the evaluation value calculation unit 25 acquires the sound pressure level of the reference sound 6 at that time from the control unit 22, and the auditory saliency level (auditory saliency level). (Evaluation value of sex) is output (S25).

  Alternatively, the evaluation value calculation unit 25 uses the reference sound 6 in the input information sequence in the predetermined time interval instead of the ratio of the input information corresponding to the reference sound 6 in the input information sequence in the predetermined time interval. A ratio of the number of input information corresponding to the number of input information corresponding to the target sound 5 may be used.

<Comparison with conventional methods>
Techniques such as Reference Non-Patent Documents 1 to 3 are known as techniques for evaluating by a method of tapping according to sound. Reference Non-Patent Document 1 presents a reference sound and a target sound to the left and right ears, respectively, and instructs to tap in accordance with the target sound. At this time, the sound pressure level at which the tap with respect to the target sound is disturbed by the reference sound is obtained. The purpose of this experiment is to measure the superiority of the left and right ears based on the difference in the sound pressure levels of the interfering sounds of the left and right ears by switching the sounds presented to the left and right ears.

  Reference Non-Patent Document 2 presents a reference sound and a target sound in the same manner as in the present invention, but instructs to tap in accordance with the target sound. Then, the presentation timing of the reference sound is gradually brought closer to the presentation timing of the target sound, and the presentation timing of the disturbing sound in which the tap on the target sound is obstructed by the reference sound is obtained. This is intended to measure simultaneity for each delay amount (time) in auditory feedback. The above techniques are common in that “tapping is performed in accordance with one of the two types of sounds, the reference sound and the target sound”, but the purpose is completely different.

  In Non-patent Document 1, in order to measure the superiority of the left and right ears, different sounds must be presented to the left and right ears, and the target sound and the reference sound can be heard in the left and right ears as in the present invention. In this case, the purpose of measuring the superiority of the left and right ears cannot be achieved. Further, Reference Non-Patent Documents 2 and 3 determine the presentation timing of an interfering sound in which a tap with respect to a target sound is obstructed by a reference sound, that is, a time difference. Since there is no clear correlation between such a time difference and auditory saliency, it is difficult to obtain an index of auditory saliency from the time difference.

In addition, by instructing any of the techniques to perform tapping in accordance with the target sound, there is a possibility that the target sound is biased to be tapped, and this bias may blur the saliency index. . On the other hand, the present invention aims to measure auditory saliency, and has a configuration in which the tapping is performed according to the reference sound and the target sound that are easy to be tapped without giving any instruction to the listener (either You can measure the saliency (if you just tap it together). Further, by providing no teaching, a more objective index value can be obtained.
(Reference Non-Patent Document 1) Tsunoda (1975) “Functional Differences Between Right- and Left-Cerebral Hemispheres Detected by the Key-Tapping Method”
(Reference Non-Patent Document 2) Aschersleben & Prinz (1997) “Delayed Auditory Feedback in Synchronization”
(Reference 3) Finney & Warren (2002) “Delayed auditory feedback and rhythmic tapping: Evidence for a critical interval shift”

Claims (9)

A sound presenting unit for presenting a sound string composed of a plurality of first sounds and a plurality of second sounds spaced apart from each other;
An input information acquisition unit for acquiring a series of beat information of a person who beats while listening to the presented sound string;
An evaluation value calculation unit that calculates an auditory saliency level based on a ratio of beat information corresponding to one of the two sounds in the series of beat information within a predetermined time interval;
When the ratio of beat information corresponding to one of the two sounds in the series of beat information within the predetermined time interval is greater than a predetermined ratio, the sound pressure level of the one sound is Make it smaller
The sound pressure level of the one sound when the ratio of the beat information corresponding to one of the two sounds in the series of beat information within the predetermined time interval is smaller than the predetermined ratio Including a control unit for increasing
The sound presenting unit controls the sound pressure controlled by the control unit until the ratio of the beat information corresponding to the one sound in the series of beat information within the predetermined time interval becomes equal to the predetermined ratio. Repeatedly presenting one sound of the above level and the other sound of a predetermined sound pressure level,
The evaluation value calculation unit, when a ratio of beat information corresponding to the one sound in the series of beat information within the predetermined time interval is equal to the predetermined ratio, a sound pressure level of the one sound The value corresponding to is output as the auditory saliency level.
Hearing Satoshiteki saliency evaluation device. Using a plurality of types of evaluation target sounds prepared in advance and a reference sound prepared in advance that is different from any of the evaluation target sounds, a plurality of evaluation target sounds and a plurality of the above reference standards separated by time intervals sound that consists of sound columns, a sound presentation section for presenting for each of the plurality of types of evaluation of the sound,
An input information acquisition unit for acquiring a series of beat information of a person who beats while listening to the presented sound string;
Based on the ratio of the beat corresponding to that beat information on Symbol reference tone of sequence information of a predetermined time interval, greater the higher the salience of the evaluated sound, saliency of the evaluated sound An evaluation value calculation unit that calculates an auditory saliency level that decreases as the value of each of the plurality of types of sounds to be evaluated;
An auditory saliency evaluation apparatus including: A auditory saliency evaluation apparatus according to claim 2 Symbol placement,
The auditory saliency level is a value in a monotonically increasing relationship with respect to the ratio of the upper SL evaluated the beat information corresponding to the sound of the series of the beat information within the predetermined time interval, the value the larger, the sound of the object of evaluation is an indication that high remote auditory saliency by the reference sound auditory saliency evaluation device. A auditory saliency evaluation apparatus according to claim 2 Symbol placement,
If the proportion of the beat corresponding to that beat information on Symbol reference tone of sequence information within the predetermined time interval is greater than a predetermined ratio, to reduce the sound pressure level of the upper Symbol reference tone,
If the proportion of the correspondingly have beat information on Symbol reference tone of sequences of said beat information within the predetermined time interval is less than the predetermined ratio, the control unit to increase the sound pressure level of the upper Symbol reference tone In addition,
The sound presentation unit, the sound pressure ratio corresponding to that beat information on Symbol reference sound to equal to the predetermined ratio, which is controlled by the control unit of the series of the beat information within the predetermined time interval level repeatedly presented over Symbol reference tone and the evaluation target sound of a given sound pressure level,
The evaluation value calculating unit, when the proportion of the beat information corresponding to the upper Symbol reference tone of sequences of said beat information within the predetermined time interval is equal to the predetermined ratio, the sound pressure level of the upper Symbol reference tone auditory saliency evaluation device for outputting a level above auditory saliency values corresponding to.   A sound presenting step for presenting a sound string composed of a plurality of first sounds and a plurality of second sounds spaced apart from each other;
  An input information acquisition step of acquiring a series of beat information of a person who beats while listening to the presented sound string;
  An evaluation value calculating step for calculating an auditory saliency level based on a ratio of beat information corresponding to one of the two sounds in the series of beat information within a predetermined time interval;
  When the ratio of beat information corresponding to one of the two sounds in the series of beat information within the predetermined time interval is greater than a predetermined ratio, the sound pressure level of the one sound is Make it smaller
  The sound pressure level of the one sound when the ratio of the beat information corresponding to one of the two sounds in the series of beat information within the predetermined time interval is smaller than the predetermined ratio Including control steps to increase
  In the sound presenting step, the sound pressure controlled in the control step until the ratio of beat information corresponding to the one sound in the series of beat information in the predetermined time interval becomes equal to the predetermined ratio. Repeatedly presenting one sound of the above level and the other sound of a predetermined sound pressure level,
  In the evaluation value calculating step, when the ratio of beat information corresponding to the one sound in the beat information series in the predetermined time interval is equal to the predetermined ratio, the sound pressure level of the one sound The value corresponding to is output as the auditory saliency level.
Auditory saliency evaluation method.   Using a plurality of types of evaluation target sounds prepared in advance and a reference sound prepared in advance that is different from any of the evaluation target sounds, a plurality of evaluation target sounds and a plurality of the above reference standards separated by time intervals A sound presenting step for presenting a sound string composed of sounds for each of the plurality of types of sounds to be evaluated;
  An input information acquisition step of acquiring a series of beat information of a person who beats while listening to the presented sound string;
  Based on the ratio of beat information corresponding to the reference sound in the beat information series in a predetermined time interval, the higher the saliency of the evaluation target sound, the greater the saliency of the evaluation target sound. An evaluation value calculating step for calculating an auditory saliency level that decreases as the value decreases for each of the plurality of types of sounds to be evaluated;
Auditory saliency evaluation method including   An auditory saliency evaluation method according to claim 6,
  The auditory saliency level is a value that is monotonically increasing with respect to the proportion of beat information corresponding to the sound to be evaluated in the series of beat information within the predetermined time interval, and the value is The larger the value, the higher the auditory saliency of the sound to be evaluated than the reference sound.
Auditory saliency evaluation method.   An auditory saliency evaluation method according to claim 6,
  When the ratio of the beat information corresponding to the reference sound in the series of beat information within the predetermined time interval is greater than a predetermined ratio, the sound pressure level of the reference sound is reduced,
  The method further includes a control step of increasing the sound pressure level of the reference sound when the ratio of the beat information corresponding to the reference sound in the series of beat information within the predetermined time interval is smaller than the predetermined ratio. ,
  In the sound presenting step, the sound pressure level controlled in the control step until the ratio of the beat information corresponding to the reference sound in the series of beat information in the predetermined time interval becomes equal to the predetermined ratio. Repeatedly presenting the reference sound and the sound to be evaluated having a predetermined sound pressure level,
  The evaluation value calculating step corresponds to the sound pressure level of the reference sound when the ratio of the beat information corresponding to the reference sound in the series of beat information within the predetermined time interval is equal to the predetermined ratio. Is output as the auditory saliency level.
Auditory saliency evaluation method. A program for causing a computer to function as the auditory saliency evaluation apparatus according to claim 1 or al 4.

Images